Feeding Ecology of Acrossocheilus Yunnanensis (Regan, 1904), a Dominant Fish in the Headwaters of the Chishui River, a Tributary of the Yangtze River

Iranian Journal of Fisheries Sciences 19(5) 2688-2704 2020 DOI: 10.22092/ijfs.2020.122630 Feeding ecology of Acrossocheilus yunnanensis (Regan, 1904), a dominant fish in the headwaters of the Chishui River, a tributary of the Yangtze River

Zhang F.1,2,3; Liu F.1; Gao X.1*; Liu H.1; Zhang X.4; Wang X.5; Cao W.1* Received: October 2017 Accepted: April 2018 Abstract Feeding ecology of Acrossocheilus yunnanensis, a dominant fish in the headwaters of the Chishui River, a tributary of the upper Yangtze River, was studied using the analysis of gut contents. From March 2015 to January 2016, a total of 543 individuals were collected and analyzed. The results showed that A. yunnanensis was an omnivorous fish mainly feeding on chlorophytes, diatoms, and aquatic insects. The trophic level was 2.69±0.62 (mean±SD), signifying A. yunnanensis as a primary or secondary predator. Dietary shifts were found among different ontogenetic stages and seasons. Specifically, young individuals fed primarily on aquatic insects and diatoms, whereas older fish fed mainly on chlorophytes. In spring, the preferred food item was aquatic insects and in other seasons, chlorophytes became the predominant prey. Diet composition showed no differences among individuals of different sex and diel periods. The feeding intensity of A. yunnanensis was not affected by diel periods, suggesting this species feeds continuously. However, its feeding intensity was significantly influenced by seasons. Pairwise comparison found that the feeding intensity was higher in spring and autumn than that in summer and winter, with minimum food intake in winter and maximum in spring. Analysis on Amundsen graph and niche breadth index indicated that A.

Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 yunnanensis might pursue an opportunistic and moderately generalized feeding strategy, which could explain why it has become the dominant fish species in our study area. Keywords: Dietary, Ontogenetic shifts, Seasonal variations, Natural reserve, Conservation

1-The Key Laboratory of Aquatic Biodiversity and Conservation of Chinese Academy of Sciences, Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan 430072, China 2-University of Chinese Academy of Sciences, Beijing 100039, China 3-College of Environmental Science and Engineering, China West Normal University, Nanchong 637009, Sichuan Province, China 4-Department of Zoology, University of British Columbia, Vancouver V6T 1Z4, Canada 5-China Three Gorges Projects Development Co., Ltd, Beijing 100038, China *Corresponding author's Email: [email protected] (Gao, X.), [email protected] (Cao, W.)

2689 Zhang et al., Feeding ecology of Acrossocheilus yunnanensis (Regan, 1904), a dominant fish in…

Introduction contrast, the headwater ecosystem The study of fish feeding habits is structure and function, especially fish essential to understand their adaptation feeding habits and adaptation mechanisms to their environments and mechanisms, have received very little to the development of conservation and attention, hindering the development of management plans (Vinyard and O' suitable conservation plans. Brien, 1976). The feeding habit of a With a length of 437 km, the Chishui species is ususlly related to its River (27°20′-8°50′ N, 104°45′-06°51′ environmental characteristics (Zander, E) is the last undammed primary 1997). However, headwaters consist of tributary of the upper Yangtze River. It many unique and highly diverse harbors approximately 160 fish species, physico-chemical environments, which and many of these are endemic to the harbor many unique species that occur upper Yangtze River (Wu et al., 2010). nowhere else in the river ecosystem As the core region of "the national (Meyer et al., 2007). Shallow and natural reserve for rare and endemic flowing water, long average annual fishes of the upper Yangtze River", the sunlight hours, together with boulders Chishui River is still well protected and and cobbles in the substratum in the plays a very important role in headwaters (Jiang et al., 2016), which biodiversity conservation (Jiang et al., are conducive to photosynthesis and 2016). algal growth (Wang et al., 2016). Acrossocheilus yunnanensis Consequently, a huge biomass of (Cyprinidae: Barbinae) is a fish species periphytic algae is always found in endemic to China, that is exclusively headwaters (Yin et al., 2013). In distributed in the upper reaches of the addition, the large inputs of Yangtze and Pearl Rivers (Ding, 1994). allochthonous organic detritus from Generally, A. yunnanensis lives in the Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 surrounding forest zones (Vannote et al., headwaters (Ding, 1994). Due to dam 1980) and the high rates of primary construction, over exploitation, productivity in un-shaded headwaters invasion of alien species and other create an environments that is rich in human activities, the population of this food for primary consumers such as species has declined dramatically in aquatic insects (Meyer et al., 2007). many rivers over the past few decades Therefore, headwaters are abundant and has even completely disappeared with periphytic algae, organic detritus from some of its original habitats (Ye et and aquatic insects, which determine al., 2015). However, A. yunnanensis is the basic structure and function in the most dominant fish species in the headwater ecosystems. headwaters of the Chishui River (Wu et Since headwaters are usually unique al., 2010). In our investigations, this and important to the whole river species accounts for 34.5 % of the local ecosystem, they have received fisheries. Therefore, why this species extensive attentions and have been became a dominant species and how it established as protected areas. By has adapted to the environment in the

Iranian Journal of Fisheries Sciences 19(5) 2020 2690

headwaters of the Chishui River have autumn: September to October 2015, attracted attention. and winter: December 2015 to January The objectives of this study were to 2016). During each sampling, the fish (1) analyze the diet composition of A. were collected by electrofishing (180 yunnanensis qualitatively and volts AC, 5 A, and 50 Hz) and quantitatively; (2) examine the effects stationary gillnets (8 m long×1.2 m of ontogenetic, seasonal, diel and high, 5 cm mesh size) at 4-h intervals sexual variations on its feeding habits; during 24-h periods (2:00, 6:00, 10:00, (3) determine its diel and seasonal 14:00, 18:00, and 22:00 h). feeding intensity; (4) evaluate its niche In the field laboratory, the standard breadth and trophic level; and (5) length (SL, 1 mm) and body weight illustrate its feeding strategies. (BW, 0.1 g) were measured immediately after capture. The gut Materials and methods length (GL, 1 mm) was measured, and Sample collection and prey analysis the gut contents were fixed in a 4 % Fish samples were captured in the formaldehyde solution for further headwaters of the Chishui River (Fig. analysis. Samples with highly digested 1). The sampling was fixed within a 7 prey were excluded from the diet km-long area, and A. yunnanensis were analysis. Sex of each fish was landed quarterly from March 2015 to determined by examination of the January 2016 (spring: March to April gonads. 2015, summer: June to July 2015,

Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021

Figure 1: Map of the study region showing sampling area (dotted box) for Acrossocheilus yunnanensis in the Chishui River of China.

2691 Zhang et al., Feeding ecology of Acrossocheilus yunnanensis (Regan, 1904), a dominant fish in…

In the laboratory, the gut contents were and 6th age: VI (n=19). Hierarchical identified to the lowest possible taxon. cluster analysis and non-metric The utmost care was given to the multidimensional scaling (NMDS) identification of even small fragments based on the Bray-Curtis similarity to minimize the underestimation of index and the % W data were conducted small and soft prey. The food items to group the six age classes (Mitu and were examined using a dissecting Alam 2016). The % W index was microscope and a binocular microscope selected because it can overcome the and then counted and weighed to the problems that digestion poses for nearest 0.0001 g. enumerating prey items (White et al. 2004). Then, the similarity percentage Data analysis (SIMPER) routine was used to assess To assess whether the number of fish the contribution of each prey to the samples analyzed was sufficient to dissimilarity observed between groups. describe the diet with respect to total To evaluate possible seasonal dietary samples, seasons, size groups, and variation, the samples were analyzed sexes, cumulative prey curves (Ferry with respect to the season. Diel dietary and Cailliet 1996) were constructed variation was analyzed by sorting the using EstimateS 9.1.0 samples into six classes according to (http://purl.oclc.org/estimates). The sampling time, namely, 2:00, 6:00, slope of the linear regression (b) of the 10:00, 14:00, 18:00, and 22:00. Finally, last five subsamples was utilized for to assess possible dietary differences this assessment, where b≤0.05 signified based on sex, the samples were divided sufficient samples for the dietary into an unidentified group (n=5), a description (Brown et al., 2012). female group (n=212), and a male The contribution of each prey to the group (n=163). Seasonal, diel, and Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 diet was quantified using several sexual diet variations were tested by indices: the average percent abundance analysis of similarities (ANOSIM) of number and weight (% AN, % AW), through the Bray-Curtis similarity the percent prey-specific abundance by matrix based on % W data. number and weight (% PN, % PW), the The feeding intensity related to diel percent frequency of occurrence (% period and season was determined by FO), and the prey-specific index of the gut fullness index (GFI), which was relative importance (PSIRI). Brown et expressed as 100%×(gut content al. (2012) have provided detailed weight/body weight) (Grabowska et al., formulas 2009). Given that the GFI was not To investigate possible ontogenetic normally distributed (Shapiro-Wilks shifts in diet, the samples were divided test, p<0.05), the variations of feeding into six size classes according to age intensity were tested using non- (Zhao et al., 2009): 1st age: I (n=11), 2nd parametric Kruskal-Wallis H test, age: II (n=66), 3rd age: III (n=51), 4th followed by Mann-Whitney U tests for age: IV (n=123), 5th age: V (n=110), pairwise comparisons.

Iranian Journal of Fisheries Sciences 19(5) 2020 2692

The niche breadth (BN) was measured All the statistical analyzes were using the standardized Levins index conducted in PRIMER 5 and SPSS 20 (Levins, 1968; Hurlbert, 1978). The at the significance level of 0.05. The Shannon-Wiener diversity index (H') images were performed by Origin pro (Shannon, 1948) was used to examine version 8.0. feeding diversity. Then, based on the % W of each prey, the trophic level (TL) Results and its variation in relation to season A total of 543 individuals of A. and ontogenetic group were calculated yunnanensis were collected and according to the formula proposed by examined, with the SL ranging from 55 Cortés (1999). The TL of animal prey to 253 (124.3±30.4, mean±SD) mm, was obtained from the research of Ebert and the BW ranging from 3.5 to 339.5 and Bizzarro (2007), and the vegetable (41.8±32.4) g (Table 1). Among the prey was defined as 1.0. samples, 43 with empty guts and 380 Finally, the feeding strategy was containing prey (gut fullness equal to or described by the Amundsen graphical greater than 20%) were used for diet method (Amundsen et al., 1996). The analysis (Table 1). All nine cumulative distribution of prey along the diagonals prey curves reached an asymptote (b< and axes of the diagram provides 0.05) (Fig. 2); therefore, the number of information about the feeding strategy, samples was considered sufficient to niche width contribution, and prey describe the diet. importance.

Table 1: Body size (standard length and body weight) and number of Acrossocheilus yunnanensis during the entire project. N represents the total number of samples for each class, and n represents the number of guts analyzed. Standard length (mm) Body weight (g)

Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 Classification N n Range Mean ± SD Range Mean ± SD Season Spring 85-188 136.0 ± 19.6 11.7-152.1 52.5 ± 24.5 61 37

Summer 77-190 129.1 ± 23.9 8.1-129.1 44.8 ± 25.0 153 99

Autumn 55-188 104.1 ± 23.9 3.5-113.2 23.5 ± 19.1 171 134

Winter 68-253 136.9 ± 34.6 5.2-339.5 54.5 ± 42.7 158 110

Diel period 2:00 84-204 141.0 ± 28.5 11.2-156.2 57.4 ± 30.8 89 72

6:00 55-253 134.7 ± 32.7 3.5-339.5 53.3 ± 41.4 144 88

10:00 79-178 118.2 ± 20.5 9.4-73.6 30.3 ± 15.1 62 46

14:00 70-177 101.5 ± 19.5 6.9-113.3 22.0 ± 15.5 90 67

18:00 68-163 105.4 ± 22.8 5.2-77.8 25.5 ± 18.6 37 21

22:00 63-194 125.2 ± 28.3 5.3-152.1 42.2 ± 28.9 121 86

Total 55-253 124.3 ± 30.4 3.5-339.5 41.8 ± 32.4 543 380

2693 Zhang et al., Feeding ecology of Acrossocheilus yunnanensis (Regan, 1904), a dominant fish in…

Figure 2: Cumulative prey curves (solid lines) and SD (dotted lines) for (A) total, (B) spring, (C) summer, (D) autumn, (E) winter, (F) YAG (SL≤110 mm), (G) OAG (SL>110 mm), (H) female, and (I) male samples.

Diet composition (PSIRI=41.30%), of which Spirogyra The diet of A. yunnanensis contained a (one of the filamentous algae) was the wide variety of algae, plants and animal most important component. The second prey (Table 2). A total of 93 different most important prey was diatoms food taxa belonging to seven main prey (PSIRI=28.80%), and the third was

Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 categories (diatoms, chlorophytes, other aquatic insects (PSIRI=21.67%). vegetable prey, aquatic insects, According to the identified aquatic mollusca, other invertebrates, and insects, Ephemeroptera was the most remains) were identified (Table 2). The important prey, followed by Trichoptera most important prey was chlorophytes and Chironomidae larvae (Table 2).

Table 2: Diet composition of Acrossocheilus yunnanensis. Diet indices include average percent number (% AN), average percent weight (% AW), percent frequency of occurrence (% FO), percent prey-specific number (% PN), percent prey-specific weight (% PW), and prey-specific index of relative importance (PSIRI); * represents values < 0.01. Prey % AN % AW % FO % PN % PW PSIRI Diatoms 45.95 11.65 98.68 46.56 11.80 28.80 Melosira 15.44 4.93 87.11 17.73 5.66 10.19 Navicula 7.01 2.15 89.21 7.85 2.41 4.58 Nitzschia 1.60 0.21 68.95 2.32 0.31 0.91 Cymbella 2.38 0.24 72.37 3.29 0.33 1.31 Gomphonema 6.66 0.63 91.58 7.27 0.69 3.65

Iranian Journal of Fisheries Sciences 19(5) 2020 2694

Table 2 continued: Synedra 0.81 0.34 56.05 1.45 0.60 0.57 Achnanthes 1.61 0.26 68.68 2.35 0.38 0.94 Diatoma 2.46 0.67 71.84 3.42 0.93 1.56 Rhoicosphenia 0.09 * 11.84 0.73 0.02 0.04 Cocconeis 3.03 1.26 79.74 3.80 1.58 2.14 Fragilaria 0.80 0.01 58.16 1.37 0.01 0.40 Gyrosigma 0.20 0.03 27.37 0.74 0.10 0.12 Pinnularia 1.19 0.27 63.95 1.85 0.42 0.73 Cyclotella 0.38 0.03 46.05 0.83 0.07 0.21 Epithemia 0.11 0.05 31.32 0.36 0.16 0.08 Surirella 1.02 0.15 49.74 2.04 0.30 0.58 Frustulia 0.56 0.31 47.37 1.19 0.66 0.44 Diploneis 0.11 0.03 24.21 0.45 0.13 0.07 Didymosphenia 0.07 0.02 15.53 0.47 0.13 0.05 Amphora 0.01 * 2.89 0.20 0.14 * Cymatopleura 0.42 0.05 10.26 4.10 0.47 0.23 Eunotia * * 0.53 0.18 0.06 * Rhizosolenia * * 0.53 0.17 0.07 * Chlorophytes 51.33 31.26 76.84 66.80 40.68 41.30 Cosmarium 0.05 * 3.95 1.15 * 0.02 Mougeotia 0.01 * 1.58 0.41 0.16 * Scenedesmus 0.01 * 0.26 2.27 * * Oedogonium 0.01 * 0.79 1.74 0.01 0.01 Chlorella 0.47 * 19.74 2.37 * 0.23 Ankistrodesmus 0.32 * 8.68 3.63 * 0.16 Closterium 0.09 * 3.16 2.75 0.15 0.05 Actinastrum 0.14 * 0.26 52.93 * 0.07 Crucigenia 0.01 * 0.79 1.02 * * Spirogyra 49.91 31.21 63.42 78.70 49.20 40.56 Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 Cladophora 0.33 0.05 1.05 31.09 4.61 0.19 Other vegetable prey 1.52 8.24 70.00 2.17 11.78 4.88 Cyanophytes 1.14 0.06 33.16 3.45 0.18 0.60 Oscillatoria 0.24 0.05 19.74 1.22 0.25 0.15 Anabeana 0.09 * 3.16 2.73 0.01 0.04 Merismopedia 0.01 * 1.05 1.11 0.01 0.01 Spirulina 0.13 * 2.89 4.62 0.02 0.07 Phormidium 0.56 0.01 10.53 5.36 0.08 0.29 Microcystis 0.10 * 0.26 38.20 * 0.05 Aphanothece 0.01 * 0.26 2.33 * * Chroococcus * * 0.26 0.36 * * Dinoflagellates 0.15 0.01 18.95 0.79 0.06 0.08 Gymnodinium 0.12 * 17.37 0.69 0.03 0.06 Peridinium 0.03 0.01 3.16 0.95 0.19 0.02 Euglenophytes 0.10 * 0.26 36.47 * 0.05 Trachelomonas 0.10 * 0.26 36.47 * 0.05 Rhodophytes * 3.37 14.21 0.01 23.69 1.68 Lemanea sinica * 3.37 14.21 0.01 23.69 1.68

2695 Zhang et al., Feeding ecology of Acrossocheilus yunnanensis (Regan, 1904), a dominant fish in…

Table 2 continued: Organic detritus 0.12 2.82 37.11 0.31 7.59 1.47 Other plant material 0.01 1.99 9.21 0.09 21.61 1.00 Vascular plants * 0.08 1.32 0.01 6.39 0.04 Plant seeds 0.01 1.91 7.89 0.10 24.15 0.96 Aquatic insects 0.76 42.58 80.26 0.95 53.05 21.67 Odonata * 0.01 0.26 * 3.99 0.01 Gomphidae larvae * 0.01 0.26 * 3.99 0.01 Plectoptera * 0.35 1.05 0.03 33.09 0.17 Perlidae * 0.35 1.05 0.03 33.09 0.17 Trichoptera * 2.53 15.79 0.02 16.00 1.26 Ephemeroptera 0.34 32.86 71.05 0.48 46.24 16.60 Diptera * 2.10 24.74 0.01 8.48 1.05 Chironomidae larvae * 1.77 22.89 0.01 7.73 0.89 Psychodidae * 0.03 2.63 0.01 1.29 0.02 Tipulidae * 0.25 0.26 * 94.59 0.12 Tabanidae * 0.04 0.53 * 8.52 0.02 Coleoptera 0.04 2.02 11.05 0.41 18.31 1.03 Dytiscidae larvae 0.04 1.33 5.26 0.85 25.22 0.69 Dytiscidae adult * 0.18 0.26 * 70.04 0.09 Hydrophilidae larvae * 0.47 4.21 * 11.06 0.23 Hydrophilidae adult * 0.05 1.58 * 2.95 0.02 Megaloptera 0.26 0.95 2.11 12.50 44.91 0.60 Sialidae 0.26 0.63 1.32 20.00 47.89 0.45 Corydalidae * 0.32 0.79 * 39.93 0.16 Hemiptera * 0.28 0.79 * 35.92 0.14 Aphelochirus * 0.03 0.26 * 13.04 0.02 Naucoris exclamationis * 0.25 0.53 * 47.36 0.12 Unidentified 0.11 1.49 10.26 1.10 14.47 0.80 * 1.56 12.63 0.01 12.39 0.78 Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 Mollusca Bivalvia * 0.76 7.89 0.01 9.60 0.38 Limnoperna lacustris * 0.71 7.11 0.01 10.05 0.36 Cuneopsis heudei * 0.04 0.79 * 5.53 0.02 Gastropoda * 0.81 5.53 0.01 14.60 0.40 Radix * 0.77 5.00 0.01 15.31 0.38 Bellamya * 0.04 0.53 0.01 7.92 0.02 Other invertebrates 0.43 0.71 26.58 1.63 2.66 0.57 Terricolous insects * 0.19 3.68 0.01 5.28 0.10 Hymenoptera * 0.19 3.42 0.01 5.46 0.09 Unidentified * 0.01 0.26 * 2.94 * Oligochaeta * 0.32 0.79 0.01 40.74 0.16 Earthworm * 0.32 0.79 0.01 40.74 0.16 Crustacea 0.02 0.19 14.21 0.16 1.31 0.10 Cladocera * 0.09 7.11 0.02 1.26 0.05 Copepoda 0.02 0.10 8.95 0.24 1.09 0.06 Rotifera * * 0.26 * 0.01 * Brachionus calyciflorus * * 0.26 * 0.01 * Protozoa 0.41 * 15.53 2.65 0.02 0.21

Iranian Journal of Fisheries Sciences 19(5) 2020 2696

Table 2 continued: Tintinnidium 0.01 * 0.53 1.75 0.09 * Stentor 0.01 * 0.26 1.92 0.07 * Oxytricha * * 0.53 0.47 0.10 * Halteria * * 0.26 0.28 * * Euplotes * * 0.53 0.24 0.03 * Chilodonella 0.27 * 10.79 2.53 0.01 0.14 Tetrahymena * * 0.26 0.83 0.01 * Coleps * * 0.79 0.09 * * Difflugia 0.04 * 3.42 1.27 * 0.02 Arcella 0.01 * 0.79 0.99 * * Actinophrys 0.04 * 0.26 14.59 * 0.02 Amoeba 0.01 * 0.79 1.25 * * Globigerina 0.02 * 1.32 1.28 * 0.01 Vorticella * * 0.26 * 0.01 * Remains * 4.00 18.68 0.01 21.39 2.00 Feather * * 0.26 * 0.02 * Woollen * * 0.26 * 0.75 * Unidentified * 3.99 18.16 0.01 22.00 2.00

Ontogenetic dietary shift aquatic insects (PSIRI=27.98%), An ontogenetic shift in the diet whereas the OAG consumed more composition was detected. The six age chlorophytes (PSIRI=49.13%) but classes can be classified in two distinct fewer aquatic insects (PSIRI=18.47%) groups through both cluster analysis and diatoms (PSIRI=23.28%) than the (complete linkage) and an NMDS YAG (Table 3). The SIMPER test ordination plot (stress=0) (Fig. 3). The indicated that the dissimilarity between two groups were defined as young age the YAG and OAG was caused mainly

Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 group (YAG: I-III, SL≤110 mm) and old by aquatic insects (33.06%), age group (OAG: IV-VI, SL>110 mm). chlorophytes (26.74%), and diatoms The data showed that the YAG mainly (15.31%). consumed diatoms (PSIRI=39.66%) and Table 3: Dietary variations of Acrossocheilus yunnanensis with season and size group. Diet indices include average percent number (% AN), average percent weight (% AW), percent frequency of occurrence (% FO), and prey-specific index of relative importance (PSIRI); * represents values<0.01.

2697 Zhang et al., Feeding ecology of Acrossocheilus yunnanensis (Regan, 1904), a dominant fish in…

Figure 3: Hierarchical cluster analysis and NMDS based on the percent of weight (% W) of the six age classes. (A) The two size groups (YAG and OAG) defined at arbitrary similarity level of 60 % are indicated (dotted line); (B) NMDS showing the ordination of the six age classes into two size groups with similar diets.

Seasonal dietary variation 40.47%) followed by diatoms The diet composition varied (PSIRI=33.76%). It consumed more conspicuously by season (ANOSIM, diatoms in winter than that in the other Global R=0.102, p<0.001). In spring, three seasons (Table 3). The other prey the predominant prey was aquatic items also varied with season (Table 3). insects (PSIRI=40.14%), followed by

Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 chlorophytes (PSIRI=27.01%) and Diel and sexual dietary variations diatoms (PSIRI=23.76%); notably, The two-way ANOSIM results showed the % FO of aquatic insects was 100% that there were no diel dietary in spring (Table 3). In summer, variations in relation to season (Global chlorophytes (PSIRI=54.76%) were the R=0.005, p>0.05) or size group (Global primary prey, whereas aquatic insects R=0.025, p>0.05). Similarly, the diet (PSIRI=12.36%) contributed the least composition did not differ between the compared to the other seasons (Table 3). sexes with respect to season (Global In autumn, the most important prey was R=0.013, p>0.05) or size group (Global chlorophytes (PSIRI=35.97%), and the R=-0.012, p>0.05). second most important prey was diatoms (PSIRI=31.57%), which had Diel and seasonal feeding intensity the highest occurrence (% FO=99.25%) Diel feeding intensity showed no compared to the other seasons (Table 3). significant difference throughout the In winter, A. yunnanensis fed entire 24-h periods (Kruskal-Wallis H predominantly on chlorophytes (PSIRI= test, p>0.05). Although no significant

Iranian Journal of Fisheries Sciences 19(5) 2020 2698

difference was found, enhanced feeding activity can be observed visually at Niche breadth, feeding diversity, trophic 14:00 and 22: 00 (Fig. 4). level and feeding strategy However, a seasonal difference in The results showed that A. yunnanensis

feeding intensity was detected has a moderate niche breadth (BN=0.38) (Kruskal-Wallis H test, p<0.001) (Fig. and high feeding diversity (H'=2.17). 5). Based on the Mann-Whitney The highest values of those indices

pairwise comparisons, the average GFI appeared in summer (BN =0.58, values were slightly higher in spring H'=2.18), whereas the lowest values

than those in autumn, although the appeared in spring (BN=0.06, H'=0.95). difference was not significant (p>0.05). In terms of ontogenetic groups, the

However, the values in both spring and OAG fish (BN =0.39, H'=2.15) had a autumn were significantly greater than greater niche breadth and feeding

those in summer and winter (p<0.05). diversity than the YAG individuals (BN Moreover, higher GFI values were =0.22, H'=1.94) (Table 4). found in summer than those in winter (p<0.05).

Table 4: Standard niche width (BN), Shannon-Wiener diversity index (H'), and trophic level (TL) of Acrossocheilus yunnanensis in relation to season, size group, and the total sample.

Classification BN H' TL (mean±SD) Spring 0.06 0.95 3.22±0.47 Summer 0.58 2.18 2.42±0.53

Autumn 0.36 1.95 2.81±0.64

Winter 0.38 2.17 2.62±0.58 Young age group 0.22 1.94 2.90±0.62 Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 Old age group 0.39 2.15 2.59±0.60

Total 0.38 2.17 2.69±0.62

Figure 4: Boxplot showing diel variation in the mean percent gut fullness index (GFI).

2699 Zhang et al., Feeding ecology of Acrossocheilus yunnanensis (Regan, 1904), a dominant fish in…

Figure 5: Boxplot showing seasonal variation in the mean percent gut fullness index (GFI).

The average TL was 2.69±0.62 in three categories (chlorophytes, (mean±SD). The highest TL was found aquatic insects, and diatoms) were in spring (TL=3.22±0.47), and the situated in the lower left, which lowest TL was observed in summer manifested that they were rare or (TL=2.42±0.53). In addition, the YAG unimportant prey. (TL=2.90±0.62) had a higher TL than the OAG (TL=2.59±0.60) (Table 4). A few prey occupied very similar positions in the Amundsen graph (Fig. 6). At the population level, all prey categories were in the lower part of the graph, signifying a generalized strategy. However, the preference of A.

Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 yunnanensis for chlorophytes, aquatic insects, and diatoms (% FO>75%) demonstrated a relatively specialized strategy. Therefore, from the perspective of the Amundsen graph and

the niche breadth (BN=0.38), A. Figure 6: Feeding strategies of Acrossocheilus yunnanensis can be considered as a yunnanensis. (A) Amundsen graph, black dot (•) represents prey category. moderate generalist predator. In terms Dia, Diatoms; Chl, Chlorophytes; O- veg, Other vegetable prey; A-ins, of niche width contribution, all prey Aquatic insects; O-inv, Other items lay on the lower right and under invertebrates; Mol, Mollusca; Rem, Remains. (B) Explanatory diagram the diagonal of the graph, for the interpretation of feeding demonstrating the individuals utilize strategy, niche width contribution and prey importance; BPC and WPC many common prey, none of which represent between-phenotype component and within-phenotype dominate the diet. Regarding prey component, respectively. importance, all prey items except those

Iranian Journal of Fisheries Sciences 19(5) 2020 2700

Discussion the third most important prey in terms Diet composition of PSIRI. This difference between the The analysis of the gut contents two studies may be attributed to revealed that A. yunnanensis is an different habitats; the specimens omnivorous feeder. The gut length (GL) investigated in the previous study were index (the ratio of GL to SL) for this collected from a different river basin. species confirmed this conclusion, with The monkey goby (Neogobius a value of 1.71±0.21 (mean±SD), which ﬂuviatilis), in the Vistula River in is in the range (1-3) for omnivores Poland exhibited significant spatial (Geevarghese, 1983). A. yunnanensis differences in diet composition has a broad trophic spectrum (Table 2); compared to those living in the largest it can take advantage of all available tributary of that river, the Bug River food resources in the environment, (Grabowska et al., 2009). In other indicating that it is an opportunistic words, A. yunnanensis is flexible in its predator. Therefore, we can conclude diet and feeds on available food that food resources are not a limiting resources. This variation may indicate a factor for population growth and strategy that considerably reduces the expansion in this species, which may be cost of seeking prey, and maximizes its one of the important reasons that it has net energy intake (Prejs and Prejs 1987). become the dominant species in headwaters of the Chishui River. Dietary variation Our results revealed that the most Ontogenetic shifts in resource use, important prey of A. yunnanensis were particularly in diet, are prevalent in fish filamentous chlorophytes, diatoms, and (Guo et al., 2013). Ontogenetic dietary aquatic insects. Tarkowska-Kukuryk shifts allow a population to share a (2013) pointed out that the diatoms and habitat by eﬀectively partitioning Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 filamentous chlorophytes are usually individuals into different feeding guilds the dominant algae groups in or ecological roles, thereby reducing periphyton. Therefore, the main prey of intra-specific competition (Wootton, A. yunnanensis were just from the 1990). Our results showed that the YAG abundant food resources in the fish consumed more aquatic insects and environment. This might be the result of diatoms, whereas the OAG individuals fish adaptation to its environment. preyed on more chlorophytes. This Ding (1994) briefly noted that A. discrepancy may be due to the yunnanensis mainly prey on morphological changes and different filamentous algae, accompanied by a metabolic requirements at different small proportion of fish and shrimp. ontogenetic stages. In this study, the GL Both Ding’s and our study indicated index was 1.55±0.23 (mean±SD) for the that filamentous algae were the primary YAG and 1.75±0.18 for the OAG. The food for A. yunnanensis. However, fish increase in GL for the OAG individuals and shrimp were not found to be prey in enhances their digestive ability by our study; instead, aquatic insects were increasing the active surface area for

2701 Zhang et al., Feeding ecology of Acrossocheilus yunnanensis (Regan, 1904), a dominant fish in…

digestion (Akin et al., 2016), thus proportion of motionless (vegetable) driving the OAG individuals to prey being consumed during the winter. consume more filamentous In fact, the biomass of macro- chlorophytes, which contain a high invertebrates in spring was higher than proportion of indigestible cellulose or that in autumn in our study area (Jiang lignin (Wootton, 1990). In contrast, to et al., 2016); and thus, the highest satisfy the demands for organ proportion of aquatic insects development and growth, the YAG fish (PSIRI=40.14%) was found in spring, must feed on prey, such as aquatic followed by autumn (PSIRI=25.40%). insects, that contain high-energy and In addition, more aquatic insects were easily digested (Barbini et al., 2010). consumed in spring, which might be Similar phenomena have been observed due to an effort to store energy for the for two species, Schizopygopsis reproductive activity that occurs in younghusbandi and S. oconnori, in summer (Ding, 1994). which younger fish tend to consume more animal prey to meet their growth Diel and seasonal feeding intensity demands (Yang et al., 2011; Ma et al., There was no apparent difference in 2014). diel feeding intensity, possibly due to The seasonal dietary variations may the large proportion (PSIRI=74.98%) of suggest that both abiotic and biotic low-energy food (vegetable prey) factors change seasonally (Wootton, consumed (Table 2). As a rule, low- 1990). In particular, those alterations energy food is evacuated faster than may directly reflect seasonal changes in high-energy food (Wootton, 1990). prey abundance or availability. The Thus, A. yunnanensis may never feel lowest proportion of mobile prey satiated and may take in food (animal prey, especially aquatic insects) continuously. S. younghusbandi, a Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 was observed in summer, probably typical fish that feeds on low-energy because macro-invertebrates were less food, feeds almost continuously and abundant in that season (Jiang et al., relies on a rapid turnover of the gut 2016). Moreover, higher water levels, contents (Yang et al., 2011). higher water velocity and reduced The feeding intensity of A. transparency caused by floods in yunnanensis displayed a clear variation summer may impede the ability of fish tendency. Our results showed that the to catch animal prey (Wootton, 1990). minimum GFI appeared in winter, Despite the high biomass of macro- possibly because the low water invertebrates in winter, the feeding temperature (10.2 °C) decreased the activity of the fish decreased feeding activity and digestion rate. The dramatically due to the low water maximum GFI was found in spring and temperature (10.2 °C) (Wootton, 1990; could be ascribed to several factors, Jiang et al., 2016). Therefore, it is not including (1) appropriate physical surprising to observe a relative lower environment factors, (2) a high proportion of aquatic insects and higher abundance of prey items, (3) a need to

Iranian Journal of Fisheries Sciences 19(5) 2020 2702

consume more food to recover vigour (CJDC-2017), the China Three Gorges after the reduced feeding period in Corporation (0799574 and 0799570), winter, and (4) a greater energy the Sino BON-Inland Water Fish requirement for gonad development and Diversity Observation Network, and the to stockpile energy for the summer National Natural Science Foundation of spawning activity. The stable China (No. 31901219). hydrologic conditions and appropriate water temperature in autumn (Jiang et References al., 2016) led to the second highest GFI Akin, S., Turan, H. and Kaymak, N., value. In addition, eating more food in 2016. Does diet variation determine autumn was conducive to storing the digestive tract length of Capoeta energy for winter (Yang et al., 2011). banarescui Turan, Kottelat, Ekmekci Floods and breeding activity in summer and Imamoglu, 2006? Journal of impede food intake and give rise to Applied Ichthyology, 32(5), 883– relatively lower GFI values. Olasotoca 892. et al. (2000) noted that for fish in Amundsen, P.A., Gabler, H.M. and spawning or pre-spawning periods, Staldvik, F., 1996. A new approach gonad development requires a certain to graphical analysis of feeding amount of space in the body cavity, strategy from stomach contents resulting in reduced feeding intensity. data—modification of the Costello Some species, such as S. (1990) method. Journal of Fish younghusbandi, cease feeding Biology, 48(4), 607–614. altogether during their spawning period Barbini, S.A., Scenna, L.B., Figueroa, (Yang et al., 2011). D.E., Cousseau, M.B. and De In conclusion, the current work Astarloa, J.M.D., 2010. Feeding provides detailed information on the habits of the Magellan skate: effects Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 feeding ecology of A. yunnanensis. The of sex, maturity stage, and body size findings of this research are a valuable on diet. Hydrobiologia, 641(1), 275– reference for developing management 286. rules for conserving this endemic Brown, S.C., Bizzarro, J.J., Cailliet, species and for managing the nature G.M. and Ebert, D.A., 2012. reserve river ecosystem. Further studies Breaking with tradition: redefining are suggested to focus on the niche measures for diet description with a partition of sympatric fish species in the case study of the Aleutian skate same study area. Bathyraja aleutica (Gilbert 1896). Environmental Biology of Fishes, Acknowledgements 95(1), 3–20. We are grateful to Dr. Teng Wang and Cortés, E., 1999. Standardized diet Zheng Gong for their assistance in compositions and trophic levels of revising the manuscript. Funding sharks. ICES Journal of Marine support was provided by the Ministry of Science: Journal du Conseil, 56(5), Agriculture and Rural Affairs of China 707–717.

2703 Zhang et al., Feeding ecology of Acrossocheilus yunnanensis (Regan, 1904), a dominant fish in…

Ding, R., 1994. The fishes of Sichuan. Longitudinal and seasonal patterns Chengdu, Sichuan, China: Sichuan of macroinvertebrate communities in Publishing House of Science and a large undammed river system in Technology, pp. 322-327 (In Southwest China. Quaternary Chinese). International, 440, 1–12. Ebert, D.A. and Bizzarro, J.J., 2007. Levins, R., 1968. Evolution in Standardized diet compositions and changing environments: some trophic levels of skates theoretical explorations. Princeton (Chondrichthyes: Rajiformes: University Press. pp. 39-65. Rajoidei). Environmental Biology of Ma, B., Xie, C., Huo, B. and Yang, Fishes, 80(2-3), 221–237. X., 2014. Feeding habits of Ferry, L. and Cailliet, G., Year., 1996. Schizothorax oconnori Lloyd, 1908 Published sample size and data in the Yarlung Zangbo River, Tibet. analysis: are we characterizing and Journal of Applied Ichthyology, comparing diet properly? In 30(2), 286–293. ‘Feeding Ecology and Nutrition in Meyer, J.L., Strayer, D.L., Wallace, Fish’.(Eds D. MacKinley and K. J.B., Eggert, S.L., Helfman, G.S. Shearer.) In: American Fisheries and Leonard, N.E., 2007. The Society Symposium: San Francisco, contribution of headwater streams to CA, 1996. pp. 70–81. biodiversity in river networks. Geevarghese, C., 1983. Morphology of JAWRA Journal of the American the alimentary tract in relation to diet Water Resources Association, 43(1), among gobioid fishes. Journal of 86–103. Natural History, 17(5), 731–741. Mitu, N. and Alam, M., 2016. Feeding Grabowska, J., Grabowski, M. and ecology of a bagrid catfish, Mystus Kostecka, A., 2009. Diet and tengara (Hamilton, 1822) in the Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 feeding habits of monkey goby Tanore wetland of Rajshahi, (Neogobius fluviatilis) in a newly Northwestern Bangladesh. Journal invaded area. Biological Invasions, of Applied Ichthyology, 32(3), 448– 11(9), 2161–2170. 455. Guo, L., Xie, P., Deng, D., Yang, H. Olasotoca, L.I., Rauschert, M. and and Zhou, Q., 2013. Seasonal Broyer, C.D., 2000. Trophic changes in icefish diel feeding ecology of the family patterns in Lake Chaohu, a large Artedidraconidae (Pisces, shallow eutrophic lake of China. Osteichthyes) and its impact on the Iranian Journal of Fisheries eastern Weddell. Sea benthic system. Sciences, 12(3), 561–576. Marine Ecology Progress, 194(1-3), Hurlbert, S.H., 1978. The 143–158. measurement of niche overlap and Prejs, A. and Prejs, K., 1987. Feeding some relatives. Ecology, 59(1), 67– of tropical freshwater fishes: 77. seasonality in resource availability Jiang, X., Xiong, J. and Xie, Z., 2016. and resource use. Oecologia, 71(3),

Iranian Journal of Fisheries Sciences 19(5) 2020 2704

397–404. Biodiversity Science, 18(2), 162–168 Shannon, C.E., 1948. A mathematical (In Chinese with English abstract). theory of communication. Bell Yang, X., Xie, C., Ma, B., Huo, B., System Technical Journal, 27(3), Huang, H., Zhang, H. and Xu, J., 379–423. 2011. Feeding habits of Tarkowska-Kukuryk, M., 2013. Schizopygopsis younghusbandi Periphytic algae as food source for younghusbandi. Freshwater grazing chironomids in a shallow Fisheries, 41(4), 40–49 (In Chinese phytoplankton-dominated lake. with English abstract). Limnologica-Ecology and Ye, S., Lin, M., Li, L., Liu, J., Song, Management of Inland Waters, L. and Li, Z., 2015. Abundance and 43(4), 254–264. spatial variability of invasive fishes Vannote, R.L., Minshall, G.W., related to environmental factors in a Cummins, K.W., Sedell, J.R. and eutrophic Yunnan Plateau lake, Lake Cushing, C.E., 1980. The river Dianchi, southwestern China. continuum concept. Canadian Environmental Biology of Fishes, Journal of Fisheries and Aquatic 98(1), 209–224. Sciences, 37(1), 130–137. Yin, X., Zhang, Y., Qu, X. and Meng, Vinyard, G.L. and O'brien, W.J., W., 2013. Spatial community 1976. Effects of light and turbidity structure of periphyton assemblages on the reactive distance of bluegill in the Taizihe River basin. Research (Lepomis macrochirus). Journal of of Environmental Sciences, 26(5), the Fisheries Board of Canada, 502–508 (In Chinese with English 33(12), 2845–2849. abstract). Wang, H., Zhou, Y., Xia, K., Yang, R. Zander, C., 1997. The distribution and and Liu, X., 2016. Flow-disturbance feeding ecology of small-size Downloaded from jifro.ir at 11:34 +0330 on Tuesday September 28th 2021 considered simulation for algae epibenthic fish in the coastal growth in a river–lake system. Mediteranean Sea. Oceanographic Ecohydrology, 9(4), 601–609. Literature Review, 2(44), 133–133. White, W., Platell, M. and Potter, I., Zhao, H., Wang, J., Hu, F., Hu, S. 2004. Comparisons between the and Cheng, Y., 2009. Age and diets of four abundant species of Growth of Acrossocheilus elasmobranchs in a subtropical Yunanensis in the Upstream Chishui embayment: implications for River. Journal of Bijie University, resource partitioning. Marine 27(8), 77–82 (In Chinese with Biology, 144(3), 439–448. English abstract). Wootton, R., 1990. Ecology of teleost ﬁshes. Chapman and Hall. pp. 32-72. Wu, J., Zhao, H., Miao, Z., Chen, Y., Zhang, F. and Wang, J., 2010. Status and conservation of fish resources in the Chishui River.